Automatic segmentation of pulmonary vasculature in thoracic CT scans with local thresholding and airway wall removal

A system for the automatic segmentation of the pulmonary vasculature in thoracic CT scans is presented. The method is based on a vesselness filter and includes a local thresholding procedure to accurately segment vessels of varying diameters. The output of an automatic segmentation of the airways is used to remove false positive detections in the airway walls. The algorithm is tested with a quantitative evaluation framework based on manual classification of well-dispersed local maxima and random points on ten axial sections in a scan. The algorithm has been applied to ten low dose CT scans annotated by two observers. Results show that local thresholding and airway wall removal both improve segmentation performance and that the accuracy of the proposed method approaches the interobserver variability.

1. Introduction

The automatic segmentation of the pulmonary vasculature is an essential preprocessing step for many algorithms that analyze thoracic CT scans. An accurate segmentation can be used to exclude the vessels from further analysis, for instance to improve nodule detection algorithms. Also, the vessel tree segmentation itself can be used for the detection of pulmonary embolisms and for quantification of hypertension in which the diameter of the arteries provide an indicator for cardiovascular risk.